!74 Check model_version and device_target in CRNN and raise error for unmatch

Merge pull request !74 from chenhaozhe/fix-crnn-version-dismatch

official/cv/crnn/README.md

@@ -18,6 +18,7 @@
             - [Distributed Training](#distributed-training)
     - [Evaluation Process](#evaluation-process)
         - [Evaluation](#evaluation)
+        - [Evaluation while training](#evaluation-while-training)
     - [Inference Process](#inference-process)
         - [Export MindIR](#export-mindir)
         - [Infer on Ascend310](#infer-on-ascend310)
@@ -43,7 +44,7 @@ CRNN use a vgg16 structure for feature extraction, the appending with two-layer
 We provide 2 versions of network using different ways to transfer the hidden size to class numbers. You could choose different version by modifying the `model_version` in config yaml.
 
 - V1 using an full connection after the RNN parts.
-- V2 change the output feature size of the last RNN, to output a feature with the same size of class numbers.
+- V2 change the output feature size of the last RNN, to output a feature with the same size of class numbers. V2 also switch to the builtin `LSTM` cell instead of operator `DynamicRNN`, which make the model being supported both on GPU and Ascend.
 
 ## [Dataset](#content)
 

official/cv/crnn/src/model_utils/config.py

@@ -122,9 +122,9 @@ def get_config():
                         help='Config file path')
     path_args, _ = parser.parse_known_args()
     default, helper, choices = parse_yaml(path_args.config_path)
-    pprint(default)
     args = parse_cli_to_yaml(parser=parser, cfg=default, helper=helper, choices=choices, cfg_path=path_args.config_path)
     final_config = merge(args, default)
+    pprint(final_config)
     return Config(final_config)
 
 config = get_config()

official/cv/crnn/train.py

@@ -55,6 +55,9 @@ def train():
         device_id = get_device_id()
         context.set_context(device_id=device_id)
 
+    if config.model_version == 'V1' and config.device_target != 'Ascend':
+        raise ValueError("model version V1 is only supported on Ascend, pls check the config.")
+
     # lr_scale = 1
     if config.run_distribute:
         if config.device_target == 'Ascend':